Re: [PATCH v2] pinctrl: samsung: add irq_set_affinity() for non wake up external gpio interrupt

[Krzysztof Kozlowski <krzk@xxxxxxxxxx>]

On 29/11/2023 09:46, Youngmin Nam wrote:
>>> I couldn't find out a pin for the test on E850-96 board yet.
>>> We can test if there is a usage of *Non" Wake up External Interrupt of GPIO
>>> on E850-96 board.
>>>
>>> Do you have any idea ?
>>
>> Please test on any upstream platform or upstream your existing platform.
>> I hesitate to take this change because I don't trust Samsung that this
>> was tested on mainline kernel. OK, for sure 100% it was not tested on
>> mainline, but I am afraid that differences were far beyond just missing
>> platforms. Therefore the issue might or might not exist at all. Maybe
>> issue is caused by other Samsung non-upstreamed code.
>>
>> Best regards,
>> Krzysztof
> 
> Sure. Let me find how to test on upstreamed device like E850-96 board.

There are many reasons why companies using Linux for their products
should be involved in upstreaming their devices.

The one visible from this conversation: Whatever technical debt you
have, it will be only growing because upstream might not even take
simple patches from you, until you start contributing with the rest.
Samsung's out-of-tree kernels are so far away from the upstream, that
basically we might feel that contributions from Samsung are not
addressing real problems. This will affect your Android trees due to GKI.

That's one more argument to talk to with your managers why staying away
from the upstream is not the best idea.

Second argument is look at your competitor: Qualcomm, one of the most
active upstreamers of SoC code doing awesome job.

Best regards,
Krzysztof